Comorbidity of Alzheimer's disease (AD) and Major Depressive Disorder (MDD) is frequent but unexplained by common genetic variants. Members of the Accelerating Medicines Partnership-Alzheimer's Disease (AMP-AD) program have exhaustively profiled gene expression in multiple brain regions from AD and control subjects through multiple cohorts and then performed systems biology analyses to identify molecular networks and drivers implicated in late onset AD. VGF (non-acronymic) is one of the top ranked AD drivers conserved in multiple cohorts. We show that VGF overexpression in hippocampus reduces neuropathology and cognitive impairment in the 5xFAD mouse model of amyloidosis (Beckmann et al., under review), and VGF is already known to have a role in depression. Its AD network includes the dual-specificity phosphatases DUSP4 and DUSP6 (MAP Kinase Phosphatases 2 and 3, respectively), all reduced in level in AD, connected via their network to Amyloid Precursor Protein/Abeta and Tau, and also previously identified by our group to be part of a network that contributes to MDD in females only. Our published and preliminary studies further demonstrate that VGF levels are reduced in MDD, in hippocampus and PFC, and that VGF overexpression in these regions has antidepressant efficacy in mice. Preliminary network analysis further identifies (1) an immune module with colony stimulating factor 1 receptor (CSF1R), a protein required for adult microglial survival, as a driver down- regulated in AD plus MDD, but up-regulated in AD alone, and (2) aquaporin-4 (AQP4), a brain water channel, which is down-regulated in AD plus MDD vs AD, is expressed in astroglial endfeet, and is implicated in AD. We hypothesize that members of our identified VGF, CSF1R, and AQP4 causal networks contribute to cognitive decline, depression-like behavior, and neuropathology in mouse models and patients with AD and MDD. In Aim 1, high throughput transcriptomics, proteomics, and multiscale network molecular modeling will be carried out on dorsolateral prefrontal cortex (DLPFC) from a new cohort of AD patients with and without comorbid MDD, MDD patients without AD, and control subjects, to identify additional shared and distinct molecular mechanisms that regulate these two diseases. In Aim 2, we propose to determine the role(s) that the VGF/DUSP shared network plays in comorbid MDD plus AD, by determining the underlying pathways by which VGF, DUSP4, and DUSP6 block or delay cognitive dysfunction, depression-like behavior, and the development of neuropathology, including microglial changes, utilizing AAV-mediated overexpression strategies in APP/PS1 mice. In Aim 3, we will validate the novel subnetworks and key drivers identified in Aim 1 that differentiate AD plus MDD from AD alone. Initially, we will investigate CSF1R/immune/microglial and AQP4/astroglial network function in depression-like behavior, neuropathology, and the regulation of gene expression (transcriptomics), in APP/PS1 mice overexpressing either CSF1R or AQP4, and also for CSF1R, in APP/PS1 mice that lack TYROBP, resulting in a normalized immune module and rescued cognitive impairment.